Open AccessMicrostructure Cluster Analysis with Transfer Learning and Unsupervised Learning
Author(s) -
Andrew R. Kitahara,
Elizabeth A. Holm
Publication year - 2018
Publication title -
integrating materials and manufacturing innovation
Language(s) - English
Resource type - Journals
SCImago Journal Rank - 0.878
H-Index - 22
eISSN - 2193-9772
pISSN - 2193-9764
DOI - 10.1007/s40192-018-0116-9
Subject(s) - artificial intelligence , unsupervised learning , transfer of learning , computer science , convolutional neural network , pattern recognition (psychology) , cluster analysis , pipeline (software) , deep learning , image (mathematics) , representation (politics) , supervised learning , machine learning , artificial neural network , politics , political science , law , programming language
We apply computer vision and machine learning methods to analyze two datasets of microstructural images. A transfer learning pipeline utilizes the fully connected layer of a pre-trained convolutional neural network as the image representation. An unsupervised learning method uses the image representations to discover visually distinct clusters of images within two datasets. A minimally supervised clustering approach classifies micrographs into visually similar groups. This approach successfully classifies images both in a dataset of surface defects in steel, where the image classes are visually distinct and in a dataset of fracture surfaces that humans have difficulty classifying. We find that the unsupervised, transfer learning method gives results comparable to fully supervised, custom-built approaches.